Cytochrome P450 2B6 (CYP2B6) and constitutive androstane receptor (CAR) polymorphisms are associated with early discontinuation of efavirenz-containing regimens

Cytochrome P450 2B6 (CYP2B6) and constitutive androstane receptor

(CAR) polymorphisms are associated with early discontinuation

of efavirenz-containing regimens

Christoph Wyen

_{, Heidy Hendra}

_{, Marco Siccardi}

_{, Martin Platten}

_{, Hans Jaeger}

_{, Thomas Harrer}

_{, Stefan Esser}

_,

Johannes R. Bogner

_{, Norbert H. Brockmeyer}

_{, Bernhard Bieniek}

_{, Juergen Rockstroh}

_{, Christian Hoffmann}

_,

Albrecht Stoehr

_{, Claudia Michalik}

_{, Verena Dlugay}

_{, Alexander Jetter}

_{, Heribert Knechten}

_{, Hartwig Klinker}

_,

Adriane Skaletz-Rorowski

_{, Gerd Fa¨tkenheuer}

_{, Deirdre Egan}

_{, David J. Back}

_{and Andrew Owen}

_{* on behalf}

of the German Competence Network for HIV/AIDS†

1

Department of Internal Medicine, University of Cologne, Cologne, Germany;

Department of Pharmacology and Therapeutics, University

of Liverpool, Liverpool, UK;

_{HIV Research and Clinical Care Centre Munich, Munich, Germany;}

_{Department of Internal Medicine 3,}

University Hospital Erlangen, University of Erlangen-Nuremberg, Germany;

Department of Dermatology, University of Essen, Essen,

Germany;

_{Department of Internal Medicine, University of Munich, Munich, Germany;}

_{Department of Dermatology, Venerology and}

Allergology, Ruhr-Universita¨t Bochum, Bochum, Germany;

PraxisCityOst, Berlin, Germany;

Department of Internal Medicine, University

of Bonn, Bonn, Germany;

_{IPM Study Centre, Hamburg/University of Schleswig Holstein, Kiel, Germany;}

_{ifi-Institute for}

Interdisciplinary Medicine, Hamburg, Germany;

Clinical Trials Centre Cologne ZKS, University of Cologne, Cologne, Germany;

Institute

of Medical Statistics, Informatics and Epidemiology, University of Cologne, Cologne, Germany;

_{Department of Clinical Pharmacology}

and Toxicology and Department of Internal Medicine, University Hospital Zurich, Zurich, Switzerland;

PZB Blondelstrasse, Aachen,

Germany;

_{Department of Internal Medicine, University of Wu¨rzburg, Wu¨rzburg, Germany;}

_{Competence Network for HIV/AIDS,}

Ruhr-Universita¨t Bochum, Bochum, Germany

*Corresponding author. Tel:+44-151-794-5919; Fax: +44-151-794-5656; E-mail: aowen@liv.ac.uk †Members of the German Competence Network for HIV/AIDS are listed in the Acknowledgements section.

Received 9 February 2011; returned 19 March 2011; revised 2 June 2011; accepted 4 June 2011

Objectives: Cytochrome P450 2B6 (CYP2B6) is responsible for the metabolic clearance of efavirenz and single

nucleotide polymorphisms (SNPs) in the CYP2B6 gene are associated with efavirenz pharmacokinetics. Since the

constitutive androstane receptor (CAR) and the pregnane X receptor (PXR) correlate with CYP2B6 in liver, and a

CAR polymorphism (rs2307424) and smoking correlate with efavirenz plasma concentrations, we investigated

their association with early (,3 months) discontinuation of efavirenz therapy.

Methods: Three hundred and seventy-three patients initiating therapy with an efavirenz-based regimen were

included (278 white patients and 95 black patients; 293 male). DNA was extracted from whole blood and

genotyping for CYP2B6 (516G

T, rs3745274), CAR (540CT, rs2307424) and PXR (44477TC, rs1523130;

63396C

T, rs2472677; and 69789AG, rs763645) was conducted. Binary logistic regression using the

back-wards method was employed to assess the influence of SNPs and demographics on early discontinuation.

Results: Of the 373 patients, 131 withdrew from therapy within the first 3 months. Black ethnicity [odds ratio

(OR) ¼ 0.27; P ¼ 0.0001], CYP2B6 516TT (OR¼ 2.81; P ¼ 0.006), CAR rs2307424 CC (OR ¼1.92; P ¼ 0.007) and

smoking status (OR ¼0.45; P ¼ 0.002) were associated with discontinuation within 3 months.

Conclusions: These data indicate that genetic variability in CYP2B6 and CAR contributes to early treatment

discontinuation for efavirenz-based antiretroviral regimens. Further studies are now required to define the

clinical utility of these associations.

Keywords: pharmacogenetics, pharmacokinetics, metabolism, drug disposition

#_{The Author 2011. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.} For Permissions, please e-mail: journals.permissions@oup.com

J Antimicrob Chemother 2011; 66: 2092 – 2098

Introduction

Efavirenz treatment is associated with CNS or neuropsychiatric

side effects in a large proportion of patients (25%–70%), and

can give symptoms such as dizziness, headache, abnormal

dreams and sleep disturbance, mood changes, anxiety, dizziness,

and impaired concentration.

_{CNS side effects normally arise in}

the first days/weeks of treatment and for a small minority of

patients can be prolonged for several months or emerge after

numerous weeks of treatment,

but do not always result in

dis-continuation of therapy. Disdis-continuation of efavirenz-containing

regimens has been reported with various frequencies in different

studies, ranging from 5% to 20%, and neuropsychiatric side

effects account for the majority of discontinuation cases.

The causes of CNS and other side effects related to

efavirenz-containing regimens have not been fully characterized, but

some studies indicate that neuropsychiatric side effects are

associated with high plasma concentrations, particularly in the

first weeks of treatment.

Phase I metabolism of efavirenz is predominantly catalysed by

cytochrome P450 2B6 (CYP2B6), and single nucleotide

polymorph-isms (SNPs) within the CYP2B6 gene are associated with altered

hepatic CYP2B6 expression and activity.

_{The role of the}

516GT (rs3745274) SNP in the pharmacokinetics of efavirenz

has been studied extensively.

In addition to CYP2B6, a

sec-ondary route of phase I metabolism of efavirenz is catalysed by

CYP2A6 and CYP2A6 polymorphisms have also recently been

ident-ified as determinants of efavirenz plasma concentrations.

The primary metabolite of phase I metabolism, 8-hydroxy

efavirenz, then undergoes glucuronidation prior to excretion in

the urine and bile. This phase II metabolism has recently

been attributed to UDP-glucuronosyltransferase isoform 2B7

(UGT2B7)

_{and evidence is emerging that UGT2B7}

polymorph-isms are also associated with efavirenz plasma concentrations.

Hence, a thorough knowledge of the routes of metabolism and

elimination of efavirenz have successfully identified candidate

genes for pharmacogenetic studies.

The constitutive androstane receptor (CAR) and the pregnane

X receptor (PXR) are known to regulate a complex network of

phase I and phase II metabolic enzymes in response to

xeno-biotics. Importantly, CAR and PXR expression also correlate

with CYP2B6

_{and CYP2A6}

_{expression in liver, even in the}

absence of enzyme inducers and xenobiotics. Activators of CAR

have also been shown to induce UGT2B genes in vivo,

and so

CAR appears to play a role in the basal and inducible regulation

of the enzymes involved in efavirenz metabolism.

Therefore, we hypothesized that interindividual variability in

CAR and PXR caused, for example, by polymorphisms, would

result in additional interindividual variability in CYP2B6, CYP2A6

and UGT2B7. Single nucleotide polymorphisms have been

reported in PXR

and CAR

genes, and we have previously

shown an association of a PXR SNP (63396C

T, rs2472677)

with unboosted atazanavir plasma concentrations

_{and an}

association of a CAR SNP (540CT, rs2307424) with plasma

concentrations of efavirenz.

The aim of this study was to investigate the association of

CYP2B6 (516G

T, rs3745274), CYP2A6 (CYP2A6*9B, rs8192726),

UGT2B7 (735A

G, rs28365062; and 802TC, rs7439366),

CAR (540CT, rs2307424) and PXR (44477TC, rs1523130;

63396CT,

rs2472677;

and

69789AG,

rs763645)

polymorphisms with discontinuation within 3 months of initiating

efavirenz-containing regimens. In addition, the associations with

demographic factors, including ethnicity, smoking habits and

gender, were also investigated.

Patients and methods

Patients

A total of 373 patients who were on stable efavirenz-containing highly active antiretroviral therapy (HAART) for ≥3 months (n¼242; control group) or who discontinued efavirenz-containing HAART within 3 months (n¼131; early discontinuation group) were included in this cohort study.

We included patients of the cohort of the German Competence Network for HIV/AIDS (KompNet Cohort) with documented stable efavirenz treatment for ≥3 months and compared with those who discontinued efavirenz-containing HAART within 3 months. A further inclusion criterion was the availability of already stored EDTA samples. All concomitant medications were documented and patients were excluded from the analysis if they received drugs with known drug interactions. No patients were receiving anti-TB treatments.

Out of 480 patients with stable efavirenz-containing HAART identified as a potential control group, stored EDTA samples were available from 255 patients. In 13 of these patients, genotyping for one or more SNP was not successful. Therefore, 242 patients on stable therapy were included in the final analysis. The median duration of efavirenz therapy in this group was 48 months (range 5– 99 months). Two hundred and thirteen patients were identified with early discontinuation of efavirenz-containing HAART, therefore fulfilling the inclusion criteria. In 131 patients, EDTA samples were available and were therefore included in the analysis. The median duration of efavirenz exposure in this group was 1.1 months (range 0.2– 3 months).

Whole blood was provided by the KompNet Cohort. In this prospective multicentre, Germany-wide cohort, semi-annual follow-up visits are documented, and clinical and demographic data are collected. EDTA samples are taken at enrolment and 3 years afterwards; serum samples are collected at every follow-up visit. Demographic and clinical data were collected on age, gender, weight, height, adherence (evalu-ated through pharmacy reports), ethnicity and qualitative smoking status.32 Ethical approval was granted by the ethics committee of the Ruhr-Universita¨t Bochum, Germany, and local ethics committee approval and informed consent were obtained at each site.

Genotyping

Total genomic DNA was isolated using the QIAamp DNA mini kit accord-ing to the manufacturer’s instructions. Followaccord-ing extraction, purity was assessed by comparing the A260 and A280 ratio. DNA was quantified

using the PicoGreenw _{dsDNA Quantitation Reagent (Molecular Probes,}

CA, USA) and normalized to 20 ng/mL. For CYP2B6, pre-amplification was first conducted to discriminate from the CYP2B6 pseudogene (CYP2B7) by modification of previously reported methods.11_Genotyping

for CYP2B6 (516GT), CYP2A6 (*9B), UGT2B7 (735AG and 802TC), CAR (540CT) and PXR (44477TC, 63396CT and 69789AG) was then performed by real-time PCR allelic discrimination using stan-dard methodology (958C for 15 min, followed by 40 cycles of 958C for 15 s and 608C for 1 min) in a DNA Engine Opticonw _{2 system (MJ}

Research Inc., USA). For CYP2B6 516GT (rs3745274), CYP2A6*9B (rs8192726), CAR 540CT (rs2307424), PXR 44477TC (rs1523130) and PXR 7635AG (rs6785049), Applied Biosystems pre-validated assays were utilized. The assay IDs were C__7817765_60, C__29560333_20, C__25746794_20, C__9152783_20 and C__29280426_10, respectively. For UGT2B7 735AG (rs28365062) the

forward primer, reverse primer, A probe and G probe were 5′ -CCTAAAG-TAATTATCTTGTGTCATCCACCTT-3′, 5′-CGTCAGCTTTCCCCATTGTCT-3′, 5′ -ACCCACTACATTATCTGT-VIC and 5′-CCCACTACGTTATCTGT-FAM, respect-ively. For UGT2B7 802TC (rs7439366) the forward primer, reverse primer, C probe and T probe were 5′-CTGACGTATGGCTTATTCGAAACTC-3′, 5′-TGGAGTCCTCCAACAAAATCAACAT-3′, 5′-AGTGGATGAGGAAACTT-VIC and 5′-AGAGTGGATAAGGAAACTT-FAM, respectively. For PXR 63396CT (rs2472677) the forward primer, reverse primer, C probe and T probe were 5′-GCACAAACATTTTCAATTTCAATGAAGTTCA-3′, 5′ -CATTCGGAAGACT-TATTCTATTCCTGTCT-3′, 5′-CCATATTTTTTCTGATTAAA-VIC and 5′-CCATA TTTTTTTTGATTAAA-FAM, respectively.

Statistical analysis

All data are given as frequency number (percentage), unless otherwise stated. Genotypes were tested for Hardy–Weinberg equilibrium by x2

test of observed versus predicted (from allele frequency) genotype frequencies. Statistical significance was assessed using the x2 _test,

Cramer’s V-test or the linear by linear test for differences in qualitative variables. Univariate and multivariate associations were assessed with SPSS v16 (SPSS Inc., Chicago, IL, USA) using backward logistic regression. Continuous variables are expressed as median [interquartile range (IQR)]. A statistical trend was defined as a P value ,0.1. Statistical significance was set at P,0.05.

Results

This was a cohort study and all patients were recruited between

1 January 2006 and 1 October 2009. Two hundred and forty-two

patients, who were on stable efavirenz-containing HAART for

≥3 months, were included in the control group and 131 patients

were included in the early discontinuation group (discontinuation

of efavirenz-containing HAART within 3 months). Treatment was

discontinued for unknown reasons in 55 patients (42%), for

self-reported CNS-related side effects in 72 cases (55%), virological

failure in 1 case (1%) and elevated liver enzymes or other

toxici-ties in 4 cases (3%). No documented cases of depression were

encountered in these patients. The baseline characteristics of

the patients entered in the study are presented in Table

1

.

The backbone therapy was comparable between the

discon-tinuation group and the control group. Preferential nucleoside

reverse transcriptase inhibitor (NRTI)-based backbone therapies

consisting either of tenofovir/emtricitabine or

lamivudine/zidovu-dine were used. The remaining patients were receiving

alterna-tive backbone or NRTI-free therapies. Within the early

discontinuation group, 42 of 131 patients were taking

lamivudine/zidovudine (or zidovudine) (32%) and 59 of 131

patients were taking tenofovir/emtricitabine (45%). Within the

control group, 63 of 242 patients were taking

lamivudine/zidovu-dine (26%) and 114 of 242 patients were receiving tenofovir/

emtricitabine (47%).

Association of demographics and backbone with early

(<3 months) treatment discontinuation

A higher proportion of white patients (39.2%; 109 out of 278)

dis-continued efavirenz-containing HAART within 3 months compared

with black patients (23.2%; 22 out of 95; P ¼ 0.005). Smokers

exhibited a lower frequency of discontinuation (30.9%; 73 out of

236) compared with non-smokers (42.3%; 58 out of 137;

P ¼ 0.026). No other associations (P,0.05) or trends (P,0.10)

were evident with demographic factors, such as age, sex,

weight, height or body mass index (BMI), even when stratified

for ethnicity. Two hundred and seventy-eight patients out of 373

had zidovudine, zidovudine/lamivudine or tenofovir/emtricitabine

as backbone therapy. The frequency of zidovudine/lamivudine (or

zidovudine) administration did not differ between patients who

discontinued (32%; 42 out of 131) and patients who were stable

on efavirenz (26%; 63 out of 242; P ¼0.21). Similarly, the frequency

of tenofovir/emtricitabine administration did not differ between

patients who discontinued (45%; 59 out of 131) and patients

who were stable on efavirenz (47%; 114 out of 242; P¼ 0.71).

Allele frequencies according to ethnicity

The frequencies of the allelic variants in white and black patients

are summarized in Table

2

. Differences in the frequencies of the

variant alleles between ethnicity groups (black ethnicity as

refer-ence) were evident for CAR rs2307424C

T [odds ratio

(OR) ¼ 2.69; P ¼0.0001], CYP2B6 516GT (OR¼0.73; P¼0.087),

CYP2B6 983TC (OR¼0; NA), UGT2B7 802TC (OR¼2.4;

P ¼ 0.0001),

PXR

7635A

G (OR¼0.14; P¼0.0001), PXR

44477T

C (OR¼7.38; P¼0.0001) and PXR 63396CT

(OR ¼1.77; P ¼ 0.009).

Association of genetic factors with early (<3 months)

treatment discontinuation

When considering the entire population, CYP2B6 516G

T was

statistically associated with early (,3 months) treatment

Table 1. Baseline characteristics of patients with stable efavirenz therapy and with early treatment discontinuation of efavirenz

On therapy .3 months On therapy ,3 months Total

Number of patients 242 131 373

Male gender 189 104 293 (78.6%)

White ethnicity 169 109 278 (74.5%)

Median age, years (IQR) 45 (40–51) 47 (41–53) 46 (26–82) Median BMI, kg/m2_{(IQR) 23.5 (20.5– 27.5) 23.5 (20.5–28) 24 (21–28)}

Smoking 163 (67.4%) 73 (55.7%) 236 (63.3%)

Median years since diagnosis (range) 6 (1 –12) 6 (1–12) 6 (1– 12) Viral load, copies/mL (range) ,50 (,50–8×105) 325 (,50–1×106) ,50 (,50–1×106) CD4 count, cells/mm3_{(range) 449 (12– 3233) 325 (0–1260) 407 (0 –3233)}

discontinuation. The discontinuation rate was 33.5% (57 out of

170) in the 516GG group compared with 32.5% (55 out of 169)

in the 516GT group and 44.1% (15 out of 34) in the 516TT group

(P ¼ 0.03). Patients characterized by T homozygosity had a

higher probability of early treatment discontinuation (55.9%

versus 33%; P ¼ 0.008). A trend towards a higher frequency of

dis-continuation among patients characterized by CAR rs2307424CC

was also observed, with 38.9% (74 out of 190) of CC patients

versus 30% (54 out of 180) of CT/TT patients discontinuing

therapy before three months (P ¼0.071; CAR genotype data

were unavailable for 3 patients). Patients with the TT genotype

for PXR 63396C

T had a trend towards a higher frequency of

dis-continuation (41.5%; 51 out 123) compared with patients with

the CT/CC genotype (32%; 80 out of 250; P ¼0.072). Similarly,

for UGT2B7 802T

C, patients with the TC or CC genotype had a

trend towards higher frequency of discontinuation (37.5%; 101

out of 269) compared with patients with the TT genotype

(28.2%; 29 out of 103; P ¼ 0.089; the UGT2B7 802 genotype was

unavailable for 1 patient). No association (P,0.05) or trend

(P,0.10) was observed for other polymorphisms.

When considering only white patients (n¼ 278), patients with

CYP2B6 516TT had a higher discontinuation rate (58.3%; 14 out

of 24) compared with patients with CYP2B6 516GG/GT (95 out of

254 patients; P ¼ 0.045). Moreover, white patients with CAR

rs2307424CC had a frequency of early discontinuation of

46.3% (56 out of 121) versus 32.5% (50 out of 154) for CT/TT

patients (P ¼ 0.019). In black patients, the CYP2B6 516TT

geno-type was associated with a higher discontinuation rate (50%;

5 out of 10) than CYP2B6 516GG/GT (20%; 17 out of 85;

P ¼ 0.033). No other associations (P,0.05) or trends (P,0.10)

were evident in either white or black patients.

Multivariate analysis by binary logistic regression

Multivariate logistic regression in the entire cohort confirmed the

independent association of CYP2B6 516 and CAR rs2307424

gen-otypes with early (,3 months) treatment discontinuation. In

addition to these genetic factors, ethnicity and smoking status

were also identified as independent predictors. As shown in

Table

3

, the following parameters were independently associated

with treatment discontinuation: CYP2B6 516TT [OR¼ 2.81;

95%

confidence

interval

(CI) ¼ 1.34–5.9;

P ¼ 0.006];

CAR

rs2307424CC (OR¼ 1.92; 95% CI ¼ 1.2–3.07; P ¼ 0.007); black

ethnicity (OR¼ 0.27; 95% CI ¼0.14 –0.50; P ¼ 0.0001); and

smoking status (OR ¼ 0.45; 95% CI¼ 0.28 –0.74; P ¼ 0.002).

Table 2. Allele frequencies in patients, considering ethnicity

Polymorphism Variant allele in white patients (%) Variant allele in black patients (%) OR P value CYP2B6 516GT 167/556 (30%) 70/190 (37%) 0.73 0.087 CYP2B6 983TC 0/556 (0%) 8/190 (4%) NA NA CAR rs2307424CT 180/550 (32.7%) 29/190 (15%) 2.69 0.0001 PXR 44477TC 329/556 (59.1%) 30/190 (15.7%) 7.38 0.0001 PXR 63396CT 339/556 (60.9%) 89/190 (46.8%) 1.77 0.009 PXR 7635AG 237/556 (42.6%) 158/188 (84%) 0.14 0.0001 CYP2A6*9B 40/556 (7.8%) 8/190 (4.2%) 1.78 0.17 UGT2B7 735AG 69/556 (12.4%) 31/190 (16.3%) 0.68 0.101 UGT2B7 802TC 296/556 (53.2%) 61/188 (32.4%) 2.4 0.0001 NA, not applicable.

Table 3. Univariate and multivariate backward logistic regression for the identification of factors associated with early treatment discontinuation of efavirenz

Covariate

Univariate Multivariate

OR (95% CI) P value OR (95% CI) P value

Gender (female) 0.92 (0.55–1.56) 0.77 — Ethnicity (black) 0.47 (0.27–0.80) 0.05 0.27 (0.14–0.50) 0.0001 Age (years) 1 (0.99–1) 0.25 — Height (cm) 0.99 (0.97–1.01) 0.57 Weight (kg) 0.99 (0.98–1.01) 0.58 Smoking (non-smoker) 0.61 (0.39–0.94) 0.03 0.45 (0.28–0.74) 0.002 CYP2B6 516TT 2.56 (1.25–5.2) 0.01 2.81 (1.34–5.90) 0.006 CYP2B6 983C carrier 0.25 (0.031– 2.12) 0.21 — CAR rs2307424CC 1.49 (0.96–2.3) 0.07 1.92 (1.20–3.07) 0.007 PXR 63396C carrier 0.66 (0.43–1.04) 0.07 ND 0.112 CYP2A6*9B carrier 1.4 (0.75–2.65) 0.29 UGT2B7 735G carrier 0.75 (0.45–1.25) 0.28 UGT2B7 802C carrier 1.53 (0.93–2.5) 0.09 ND 0.22 ND, not done.

Rows with a significant P value (P,0.05) in the multivariate analysis are shown in bold. Constant of the final multivariate model¼0.74; P¼0.21.

Measure of goodness of fit was conducted using the Hosmer–Lemeshow test: x2¼ 4.8; P¼0.56.

Composite analysis for genetic and demographic

risk factors

To further confirm the association of these two polymorphisms

with

early

discontinuation,

the

CYP2B6

516T and

CAR

rs2307424C alleles were classified as discontinuation-associated

alleles. As shown in Figure

1

(a), the number of high

discontinuation-associated alleles correlated with the rate of

early discontinuation. The rate of treatment discontinuation

was 25% (4 out of 16) for patients with no

discontinuation-associated alleles, 29.6% (24 out of 81) for patients with one

discontinuation-associated allele, 32.4% (48 out of 148) for

patients with two alleles, 39.4% (43 out of 109) for patients

with three alleles and 56.2% (9 out of 16) for patients with

four alleles (P ¼ 0.023; CAR genotype data were unavailable for

3 patients).

As shown in Figure

1

(b), the number of early discontinuation

risk factors harboured by a patient [including CYP2B6 and CAR

genetics (with each allele counting as one risk factor), ethnicity,

and smoking status] was associated with early discontinuation.

The rate of early treatment discontinuation was 0% (n¼ 1) for

patients with no risk factors, 16.7% (3 out of 18) for patients

with one risk factor, 23% (16 out of 69) for patients with two

risk factors, 31% (45 out of 145) for patients with three risk

factors, 43.9% (43 out of 98) for patients with four risk factors,

51.4% (19 out of 37) for patients with five risk factors and

100% (n¼2) for patients with six risk factors (P ¼ 0.0001; CAR

genotype data were unavailable for 3 patients).

Discussion

Several studies have described an elevated frequency of CNS side

effects in patients treated with efavirenz-based regimens,

leading to treatment discontinuation primarily in the first week

of treatment.

Interindividual variability in genes involved in

the metabolism and disposition of efavirenz may influence the

onset of side effects and, therefore, therapy discontinuation.

CYP2B6 516 has been identified as a major predictor of efavirenz

plasma concentrations,

and a correlation between the

frequency of CNS side effects in white patients and

slow-metabolizer genotypes (CYP2B6 516TT and 983TT) was recently

described.

_{In the current cohort, TT homozygosity for the}

CYP2B6 516GT polymorphism was independently associated

with early treatment discontinuation.

A

novel

association

between

a

CAR

polymorphism

(rs2307424) and plasma efavirenz concentrations was recently

reported by some of the investigators involved in the present

study.

_{The discontinuation data presented here extend these}

findings to a clinical phenotype, showing significant association

of the CC genotype with early treatment discontinuation. These

findings strengthen the role of CAR in efavirenz disposition and

pharmacokinetics, presumably through its role in the regulation

of CYP2B6.

A genetic composite of these two polymorphisms (CYP2B6

516CT and CAR rs2307424) was assessed and patients with

more discontinuation-associated alleles (also the same as the

high concentration-associated alleles

_{) had a higher rate of}

dis-continuation. This finding reinforces the hypothesis that early

treatment discontinuation is associated with higher plasma

con-centrations. However, it must be noted that plasma efavirenz

concentrations are higher in black patients than in white

patients,

_{yet treatment discontinuation rates appear to be}

lower in black patients.

In the present study, white patients

had a higher frequency of early discontinuation than black

patients. Furthermore, the frequency of CYP2B6 and CAR alleles

associated with higher plasma efavirenz concentrations are

more frequent in black patients. We hypothesize that this

appar-ent paradox may be explained by interethnic variability in other

unknown genes involved in the aetiology of CNS toxicity.

0 100 90 80 70 60 50 40 30 20 10 0 1 2 3 4

Number of discontinuation-associated alleles

Patients (%) (a) P = 0.023 n = 12 n = 57 n = 100 n = 66 n = 7 n = 4 n = 24 n = 48 n = 43 n = 9 0 100 90 80 70 60 50 40 30 20 10 0 1 2 3 4 5 6

Number of discontinuation-associated risk factors

Patients (%) (b) P = 0.0001 n = 1 n = 15 n = 53 n = 100 n = 55 n = 18 n = 2 n = 3 n = 16 n = 45 n = 43 n = 19 Treatment Discontinuation Treatment Discontinuation

Figure 1. Bar chart showing the percentage of patients discontinuing treatment within 3 months and (a) number of discontinuation-associated alleles (CYP2B6 516T and CAR rs2307424C) and (b) number of discontinuation-associated risk factors (CYP2B6 516T and CAR rs2307424C, white ethnicity, and smoking).

Smoking status was also identified as being independently

associated with early treatment discontinuation. We previously

reported a trend towards an association between smoking and

efavirenz plasma concentrations in the German Competence

Network for HIV/AIDS

and in a Chilean cohort.

Interestingly,

nicotine has previously been shown to activate nuclear receptors

such as PXR and CAR,

_{and induction of CYP2B6 has been}

described.

_{Furthermore, nicotine and efavirenz have an}

overlapping elimination pathway, both being metabolized by

CYP2B6 and CYP2A6, providing another putative mechanism for

the interaction.

Nicotine exposure has also been shown to

influence blood –brain barrier permeability in vivo

and may

therefore have an effect on efavirenz diffusion in the CNS,

independently of effects on pharmacokinetics.

It must be noted that a major limitation of the current study

is that the reasons for early discontinuation were not known for

42% of the cohort. However, of those discontinuing therapy for a

known reason, 94.7% discontinued due to CNS toxicity. Other

reasons for early treatment discontinuation (e.g. virological

failure) usually lead to later (.3 months) discontinuation or

modification of therapy. For this reason, it seems likely that in

the majority of patients the reason for the treatment

discontinu-ation was CNS toxicity. Moreover, the reported associdiscontinu-ations were

also present when univariate logistic analysis was conducted for

patients who discontinued due to known CNS toxicity (n¼ 72):

CAR

rs2307424CC

(OR ¼1.86;

P ¼0.037);

CYP2B6

516TT

(OR ¼2.1; P ¼ 0.122); and black ethnicity (OR ¼ 0.42; P ¼ 0.013).

A potential additional limitation of this study is that no

infor-mation was collected with respect to their adherence to

medication.

In conclusion, our findings indicate that genetic variability in

genes involved in the metabolism and disposition of efavirenz

are determinants of early (,3 months) treatment

discontinu-ation. These data provide evidence that smoking status may

be associated with discontinuation. In vitro studies to identify

the mechanisms underpinning these associations and to identify

toxicological targets in the brain are now warranted.

Acknowledgements

First, we thank all the patients participating in our cohort for providing their data and biomaterials to the cohort. We are grateful for all the work and dedication of the documentation officers and of the heads of our documenting sites. We thank the Ruhr-Universita¨t Bochum for financial and structural encouragement. In addition, we thank the Clinical Trial Centre Cologne for its support in conducting the study. Furthermore, we would like to thank Gislea Kremer and Ellen Rund, Cologne for administrative assistance, and Winfried Siffert, Essen for collecting and providing the blood samples of the German Competence Network for HIV/AIDS.

KompNet Cohort co-ordinators

Stephan Dupke, Andreas Carganico, Axel Baumgarten (Berlin), Siegfried Koeppe, Peter Kreckel (Berlin), Elke Lauenroth-Mai, Frank Schlote, Christoph Schuler (Berlin), Matthias Freiwald, Michael Rausch (Berlin), Jo¨rg Go¨lz, Arend Moll (Berlin), Martin Zeitz (Berlin), Norbert Brockmeyer (Bochum), Martin Hower (Dortmund), Stefan Reuter (Duesseldorf), Thomas Harrer (Erlangen), Stefan Esser (Essen), Schlomo Staszewski (Frankfurt), Andreas Plettenberg (Hamburg), Stefan Fenske (Hamburg),

Thomas Buhk, Hans-Ju¨rgen Stellbrink (Hamburg), Reinhold Schmidt (Hannover), Birger Kuhlmann (Hannover), Franz Mosthaf (Karlsruhe), Ansgar Rieke (Koblenz), Stefan Scholten (Ko¨ln), Hans Jaeger, Eva Jaegel-Guedes (Mu¨nchen), Ramona Volkert, Werner Becker (Mu¨nchen), Helmut Hartl (Mu¨nchen), Antonius Mutz (Osnabru¨ck), Albrecht Ulmer, Bernhard Frietsch and Markus Mu¨ller (Stuttgart).

Funding

This work was funded by the UK Medical Research Council (G0800247), the German Competence Network for HIV/AIDS and the BMBF (Bundesministerium fu¨r Bildung und Forschung; grant: 01 KI 0771). The German Competence Network for HIV/AIDS is supported by a grant of the Federal Ministry of Education and Research (Grant No. BMBF 01 KI 0501). The Clinical Trial Center Cologne also supported the study. H. H. was supported by a scholarship of the Else Kroener Foundation.

Transparency declarations

C. W. has received honoraria for lectures and advisory boards from Abbott, Boehringer Ingelheim, Bristol-Myers Squibb, VIIV, Gilead, Pfizer and Essex. G. F. has received honoraria for lectures and advisory boards from Abbott, Boehringer Ingelheim, Bristol-Myers Squibb, Gilead, GlaxoSmithKline, Novartis, Pfizer, Roche, Schering-Plough and Tibotec. D. J. B. and A. O. have received research funding from Boehringer Ingelheim, GlaxoSmithKline, Abbott Laboratories, Pfizer, AstraZeneca, Tibotec, Merck and Roche Pharmaceuticals. All other authors: none to declare.

References

1 Fumaz CR, Tuldra A, Ferrer MJ et al. Quality of life, emotional status, and adherence of HIV-1-infected patients treated with efavirenz versus protease inhibitor-containing regimens. J Acquir Immune Defic Syndr 2002; 29: 244–53.

2 Hawkins T, Geist C, Young B et al. Comparison of neuropsychiatric side effects in an observational cohort of efavirenz- and protease inhibitor-treated patients. HIV Clin Trials 2005; 6: 187– 96.

3 Perez-Molina JA. Safety and tolerance of efavirenz in different antiretroviral regimens: results from a national multicenter prospective study in 1,033 HIV-infected patients. HIV Clin Trials 2002; 3: 279–86. 4 Fumaz CR, Munoz-Moreno JA, Molto J et al. Long-term neuropsychiatric disorders on efavirenz-based approaches: quality of life, psychologic issues, and adherence. J Acquir Immune Defic Syndr 2005; 38: 560– 5. 5 Manfredi R, Calza L, Chiodo F. Efavirenz versus nevirapine in current clinical practice: a prospective, open-label observational study. J Acquir Immune Defic Syndr 2004; 35: 492– 502.

6 Sabbatani S, Fulgaro C, Rosticelli E. [Efficacy of efavirenz in a mixed population of patients. Identification of drop-out causes]. Infez Med 2002; 10: 163–8.

7 Spire B, Carrieri P, Garzot MA et al. Factors associated with efavirenz discontinuation in a large community-based sample of patients. AIDS Care 2004; 16: 558– 64.

8 Nunez M, Gonzalez de Requena D, Gallego L et al. Higher efavirenz plasma levels correlate with development of insomnia. J Acquir Immune Defic Syndr 2001; 28: 399– 400.

9 Clifford DB, Evans S, Yang Y et al. Impact of efavirenz on neuropsychological performance and symptoms in HIV-infected individuals. Ann Intern Med 2005; 143: 714–21.

10 Gallego L, Barreiro P, del Rio R et al. Analyzing sleep abnormalities in HIV-infected patients treated with efavirenz. Clin Infect Dis 2004; 38: 430–2.

11 Lang T, Klein K, Fischer J et al. Extensive genetic polymorphism in the human CYP2B6 gene with impact on expression and function in human liver. Pharmacogenetics 2001; 11: 399– 415.

12 Haas DW, Bartlett JA, Andersen JW et al. Pharmacogenetics of nevirapine-associated hepatotoxicity: an Adult AIDS Clinical Trials Group collaboration. Clin Infect Dis 2006; 43: 783– 6.

13 Haas DW, Ribaudo HJ, Kim RB et al. Pharmacogenetics of efavirenz and central nervous system side effects: an Adult AIDS Clinical Trials Group study. AIDS 2004; 18: 2391– 400.

14 Ribaudo HJ, Haas DW, Tierney C et al. Pharmacogenetics of plasma efavirenz exposure after treatment discontinuation: an Adult AIDS Clinical Trials Group study. Clin Infect Dis 2006; 42: 401–7.

15 Ritchie MD, Haas DW, Motsinger AA et al. Drug transporter and metabolizing enzyme gene variants and nonnucleoside reverse-transcriptase inhibitor hepatotoxicity. Clin Infect Dis 2006; 43: 779– 82. 16 Rodriguez-Novoa S, Barreiro P, Rendon A et al. Influence of 516G.T polymorphisms at the gene encoding the CYP450-2B6 isoenzyme on efavirenz plasma concentrations in HIV-infected subjects. Clin Infect Dis 2005; 40: 1358– 61.

17 Rotger M, Colombo S, Furrer H et al. Influence of CYP2B6 polymorphism on plasma and intracellular concentrations and toxicity of efavirenz and nevirapine in HIV-infected patients. Pharmacogenet Genomics 2005; 15: 1– 5.

18 Tsuchiya K, Gatanaga H, Tachikawa N et al. Homozygous CYP2B6 *6 (Q172H and K262R) correlates with high plasma efavirenz concentrations in HIV-1 patients treated with standard efavirenz-containing regimens. Biochem Biophys Res Commun 2004; 319: 1322– 6.

19 Wyen C, Hendra H, Vogel M et al. Impact of CYP2B6 983T.C polymorphism on non-nucleoside reverse transcriptase inhibitor plasma concentrations in HIV-infected patients. J Antimicrob Chemother 2008; 61: 914–8.

20 Kwara A, Lartey M, Sagoe KW et al. CYP2B6, CYP2A6 and UGT2B7 genetic polymorphisms are predictors of efavirenz mid-dose concentration in HIV-infected patients. AIDS 2009; 23: 2101–6. 21 Kwara A, Lartey M, Sagoe KW et al. CYP2B6 (c.516GT) and CYP2A6 (*9B and/or *17) polymorphisms are independent predictors of efavirenz plasma concentrations in HIV-infected patients. Br J Clin Pharmacol 2009; 67: 427–36.

22 di Iulio J, Fayet A, Arab-Alameddine M et al. In vivo analysis of efavirenz metabolism in individuals with impaired CYP2A6 function. Pharmacogenet Genomics 2009; 19: 300–9.

23 Arab-Alameddine M, Di Iulio J, Buclin T et al. Pharmacogenetics-based population pharmacokinetic analysis of efavirenz in HIV-1-infected individuals. Clin Pharmacol Ther 2009; 85: 485– 94.

24 Belanger AS, Caron P, Harvey M et al. Glucuronidation of the antiretroviral drug efavirenz by UGT2B7 and an in vitro investigation of drug– drug interaction with zidovudine. Drug Metab Dispos 2009; 37: 1793– 6.

25 Chang TK, Bandiera SM, Chen J. Constitutive androstane receptor and pregnane X receptor gene expression in human liver: interindividual variability and correlation with CYP2B6 mRNA levels. Drug Metab Dispos 2003; 31: 7 –10.

26 Wortham M, Czerwinski M, He L et al. Expression of constitutive androstane receptor, hepatic nuclear factor 4 a, and P450 oxidoreductase genes determines interindividual variability in basal expression and activity of a broad scope of xenobiotic metabolism genes in the human liver. Drug Metab Dispos 2007; 35: 1700–10. 27 Shelby MK, Klaassen CD. Induction of rat UDP-glucuronosyltransferases in liver and duodenum by microsomal enzyme inducers that activate various transcriptional pathways. Drug Metab Dispos 2006; 34: 1772– 8. 28 Lamba J, Lamba V, Strom S et al. Novel single nucleotide polymorphisms in the promoter and intron 1 of human pregnane X receptor/NR1I2 and their association with CYP3A4 expression. Drug Metab Dispos 2008; 36: 169–81.

29 Jinno H, Tanaka-Kagawa T, Hanioka N et al. Identification of novel alternative splice variants of human constitutive androstane receptor and characterization of their expression in the liver. Mol Pharmacol 2004; 65: 496–502.

30 Siccardi M, D’Avolio A, Baietto L et al. Association of a single-nucleotide polymorphism in the pregnane X receptor (PXR 63396CT) with reduced concentrations of unboosted atazanavir. Clin Infect Dis 2008; 47: 1222–5.

31 Cortes C, Chaikan A, Owen A et al. Correlates of efavirenz exposure in HIV infected patients from Chile reveals novel associations with a polymorphism in the constitutive androstane receptor and smoking. In: Abstracts of the Tenth International Workshop on Clinical Pharmacology of HIV Therapy, Amsterdam, Netherlands 2009. Abstract P-04. Rev Antivir Ther 2009; Suppl.

32 Jansen K, Michalik C, Hahn M et al. The patient cohort of the German Competence Network for HIV/AIDS (KompNet): a profile. Eur J Med Res 2009; 14: 323–31.

33 Ribaudo HJ, Liu H, Schwab M et al. Effect of CYP2B6, ABCB1, and CYP3A5 polymorphisms on efavirenz pharmacokinetics and treatment response: an AIDS Clinical Trials Group study. J Infect Dis 2010; 202: 717– 22. 34 Stohr W, Back D, Dunn D et al. Factors influencing efavirenz and nevirapine plasma concentration: effect of ethnicity, weight and co-medication. Antivir Ther 2008; 13: 675– 85.

35 Schackman BR, Ribaudo HJ, Krambrink A et al. Racial differences in virologic failure associated with adherence and quality of life on efavirenz-containing regimens for initial HIV therapy: results of ACTG A5095. J Acquir Immune Defic Syndr 2007; 46: 547–54.

36 Lamba V, Yasuda K, Lamba JK et al. PXR (NR1I2): splice variants in human tissues, including brain, and identification of neurosteroids and nicotine as PXR activators. Toxicol Appl Pharmacol 2004; 199: 251– 65. 37 Washio I, Maeda M, Sugiura C et al. Cigarette smoke extract induces CYP2B6 through constitutive androstane receptor in hepatocytes. Drug Metab Dispos 2011; 39: 1 –3.

38 Yamazaki H, Inoue K, Hashimoto M et al. Roles of CYP2A6 and CYP2B6 in nicotine C-oxidation by human liver microsomes. Arch Toxicol 1999; 73: 65– 70.

39 Hawkins BT, Abbruscato TJ, Egleton RD et al. Nicotine increases in vivo blood– brain barrier permeability and alters cerebral microvascular tight junction protein distribution. Brain Res 2004; 1027: 48– 58.

40 Lockman PR, Van der Schyf CJ, Abbruscato TJ et al. Chronic nicotine exposure alters blood–brain barrier permeability and diminishes brain uptake of methyllycaconitine. J Neurochem 2005; 94: 37 –44.